Load Packages

if (!require("knitr")) {install.packages("knitr"); require("knitr")}
if (!require("BiocManager")) {install.packages("BiocManager"); require("BiocManager")}
if (!require("dplyr")) {install.packages("dplyr"); require("dplyr")}
if (!require("stringr")) {install.packages("stringr"); require("stringr")}
if (!require("Seurat")) {install.packages("Seurat"); require("Seurat")}
if (!require("sctransform")) {install.packages("sctransform"); require("sctransform")}
if (!require("glmGamPoi")) {BiocManager::install('glmGamPoi'); require("glmGamPoi")}
if (!require("patchwork")) {install.packages("patchwork"); require("patchwork")}
if (!require("ggplot2")) {install.packages("ggplot2"); require("ggplot2")}
if (!require("EnhancedVolcano")) {BiocManager::install('EnhancedVolcano'); require("EnhancedVolcano")}
if (!require("DESeq2")) {BiocManager::install('DESeq2'); require("DESeq2")}
if (!require("tidyverse")) {install.packages("tidyverse"); require("tidyverse")}
if (!require("RColorBrewer")) {install.packages("RColorBrewer"); require("RColorBrewer")}
if (!require("car")) {install.packages("car"); require("car")}
if (!require("openxlsx")) {install.packages("openxlsx"); require("openxlsx")}
if (!require("readxl")) {install.packages("readxl"); require("readxl")}
if (!require("ggrepel")) {install.packages("ggrepel"); require("ggrepel")}
if (!require("gghighlight")) {install.packages("gghighlight"); require("gghighlight")}
if (!require("ggpmisc")) {install.packages("ggpmisc"); require("ggpmisc")}
if (!require("data.table")) {install.packages("data.table"); require("data.table")}
if (!require("here")) {install.packages("here"); require("here")}
if (!require("reshape2")) {install.packages("reshape2"); require("reshape2")}
if (!require("ggpmisc")) {install.packages("ggpmisc"); require("ggpmisc")}
if (!require("devtools")) {install.packages("devtools"); require("devtools")}
if (!require("sjmisc")) {install.packages("sjmisc"); require("sjmisc")}
if (!require("gridExtra")) {install.packages("gridExtra"); require("gridExtra")}
if (!require("gplots")) {install.packages("gplots"); require("gplots")}
if (!require("ggvenn")) {install.packages("ggvenn"); require("ggvenn")}
if (!require("pheatmap")) {install.packages("pheatmap"); require("pheatmap")}
if (!require("viridis")) {install.packages("viridis"); require("viridis")}
if (!require("cowplot")) {install.packages("cowplot"); require("cowplot")}

set.seed(12345)
here()

## [1] "/Users/sblanche/Desktop/RNA-Protein Correlation"

Project Narritive

The goal of this project is to examine the relationship between RNA and protein expression.

Protein Processing

This section details how I went about deciding how to analyze each of the proteins from the western blot data. I did not factor any of the phosphorylated proteins in my analysis. For the proteins that are present in both the medulla and the cortex, like NHE3, AQP1, and NKCC2, I use a weighted average that creates a new value that accounts for the expression in both locations of the kidney. For proteins like AQP1 and ENaC which have two reported lengths, I calculated the mean of their respective values.

Narrative of workflow

After loading the data I visualize the genes of interest through a series of box blots to determine the relative amount of expression in the medulla versus in the cortex.
Next I cleaned the protein data in three steps by selecting the desired columns from the raw western blot data, calculating the weighted average for cortex and medullary proteins, and calculating the average expression for proteins with multiple methods of measurement.
I repeated a similar set of steps with the RNA data. To clean this set of data, I rotated the data frame, renamed the columns, then normalized the wild type values to one.
Once both sets of data were cleaned, I compared them to their original data frames via a bar graph for the protein data and a set of box plots for the RNA data.
Next I visualized the data with a set of scatter plots.
In order to visualize the data in an alternate method, I created a table to expressed how many of the genes/proteins expressed a basic correlation by determining how many of them shared the same patterns of expression across their RNA and protein data.
Next I examined the changed of expression between protein and RNA within groups. I did this for the transformed data, as well as to look at the data in three pairs: WT vs KO, WT vs WT+A, and KO vs KO+A. I visualized the change in expression by looking at a scatter plot of their differences as well as calculating those differences and visualizing them on a set of heat maps.
I then visualized the sum of those differences for each of those comparisons as well as the transformed data via bar graph where each point in the graph represented a sum for a gene as well as standard error bars.
As an alternative method of visualizing this information, I created a table with the sum of the deltas for each comparison.
I then ran T-Tests for each of those comparisons to determine a possible relationship between the means of pairs.
To assess the correlations between each of the pairs I compared the R squared values for each of the comparison which were calculated from the completion of a simple linear regression. These values were further visualized with a set of heat maps.
Lastly, I computed the ratio of the mean absolute values of the difference in AngII groups compared to the difference in the KO group.

Important notes

Interpretation

Load Data

#Load the PB dataset
load(here("Dataset", "Pseudobulk_cell_group.RDS"))
load(here("Dataset", "Pseudobulk_cell.RDS"))
load(here("Dataset", "Pseudobulk_group.RDS"))

# load xlsx dataset
Protein <- read_excel("Dataset/Alicia data copy.xlsx")

Visualize Genes of Interest

# gene list
gene_name <- unique(unlist(Protein$Gene))

#loop that makes a box plot of the expression of each gene in every cell population
  for (i in gene_name) {
  # Extract the specified column along with rownames
  if (i %in% colnames(Pseudobulk.cell)) {
    selected_column <- Pseudobulk.cell[, i, drop = FALSE]
    
    # Save the results to a data frame with rownames
    result_df <- data.frame(RowName = rownames(selected_column), Value = selected_column[, 1])
    
    # Ensure the RowName factor levels are consistent
    result_df$RowName <- factor(result_df$RowName, levels = c("PTS1", "PTS2", "PTS3", "dTL", "TAL", "MD", "DCT1", "DCT2", "CNT", "PC", "ICA", "ICB", "Podo", "PEC", "EC", "Fib", "Contractile", "Mes", "Macro", "Lympho", "Uro"))
    
    # Create a box plot of result_df
    f1 <- ggplot(result_df, aes(x = RowName, y = Value)) + 
      geom_boxplot() + 
      theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
      labs(title = i, x = "Cell Type", y = "Expression")
  } else {
    # Create a placeholder plot indicating the gene was not found
    f1 <- ggplot() + 
      annotate("text", x = 1, y = 1, label = paste("Gene", i, "not found in data frame"), size = 5, color = "red") + 
      theme_void() + 
      labs(title = i)
  }

  # Print plots
  print(f1)
}

Cleaning the Protein data

Step 1: Clean the raw data

#clean raw protein data
Protein.1 <- Protein %>%
  select("Gene", "Protein", "WT", "WT+A", "KO", "KO+A")  

head(Protein.1)

This is the cleaned version of the raw data.

Step 2: Calculate the weighted average for cortex and medullary proteins

# process the data according to the protein processing section
# calculate the weighted mean for medulla and cortex proteins
# Create a new column to identify labeled proteins
Protein.2 <- Protein.1 %>%
  mutate(
    IsLabeled = grepl("^m", Protein),  # Identify labeled proteins by 'm' prefix
    Protein = gsub("^m", "", Protein)  # Remove 'm' prefix to get the base protein name
  )

# Separate the labeled and unlabeled data
labeled_data <- Protein.2 %>%
  filter(IsLabeled) %>%
  select(Protein, Gene,  WT, `WT+A`, KO, `KO+A`)

unlabeled_data <- Protein.2 %>%
  filter(!IsLabeled) %>%
  select(Protein, Gene, WT, `WT+A`, KO, `KO+A`)

# Join the labeled and unlabeled data on ProteinName
combined_data <- full_join(unlabeled_data, labeled_data, by = c("Protein", "Gene"), suffix = c("_c", "_m"))

# Apply the formula (4c + m) / 5 to each of the conditions
Protein.2 <- combined_data %>%
  rowwise() %>%
  mutate(
    WT = ifelse(is.na(WT_m) | is.na(WT_c), coalesce(WT_c, WT_m), (4 * WT_c + WT_m) / 5),
    `WT+A` = ifelse(is.na(`WT+A_m`) | is.na(`WT+A_c`), coalesce(`WT+A_c`, `WT+A_m`), (4 * `WT+A_c` + `WT+A_m`) / 5),
    KO = ifelse(is.na(KO_m) | is.na(KO_c), coalesce(KO_c, KO_m), (4 * KO_c + KO_m) / 5),
    `KO+A` = ifelse(is.na(`KO+A_m`) | is.na(`KO+A_c`), coalesce(`KO+A_c`, `KO+A_m`), (4 * `KO+A_c` + `KO+A_m`) / 5)
  ) %>%
  ungroup() %>%
  select(Protein, Gene, WT, `WT+A`, KO, `KO+A`)

# round to 3 sig figs
Protein.2 <- Protein.2 %>%
  mutate_if(is.numeric, round, 3)

head(Protein.2)

This is the data after one step of processing. In this step a weighted average of the medullary and cortex proteins was calculated. To compute the weighted average, we used the formula 4c+m/5. The proteins for which we used this formula for were AQP1, NHE3, NKCC2, NKA alpha 1, NKA beta 1, HO-1, and AQP2.

Step 3: Calculate the average for proteins with multiple measurements

# process the data according to the protein processing section
# calculate mean for full length vs cleaved proteins
# Define patterns for identifying full and cleaved proteins

  # Process full proteins
  full_proteins <- Protein.2 %>%
    filter(grepl(" - FL$| - glycos$| -\\s*35 kD$| -\\s*37 kD$" , Protein)) %>%
    mutate(ProteinBase = sub(" - FL$| - glycos$| -\\s*35 kD$| -\\s*37 kD$", "", Protein))

  # Process cleaved proteins
  cleaved_proteins <- Protein.2 %>%
    filter(grepl(" - cl$| - core$| -\\s*23 kD$" , Protein)) %>%
    mutate(ProteinBase = sub(" - cl$| - core$| -\\s*23 kD$", "", Protein))
  
  # Merge full and cleaved proteins
merged_data1 <- full_proteins %>%
  inner_join(cleaved_proteins, by = c("Gene", "ProteinBase"), suffix = c("_full", "_cleaved"))

  # Calculate means for each condition
  means_data <- merged_data1 %>%
  rowwise() %>%
  mutate(
    WT = mean(c(WT_full, WT_cleaved), na.rm = TRUE),
    `WT+A` = mean(c(`WT+A_full`, `WT+A_cleaved`), na.rm = TRUE),
    KO = mean(c(KO_full, KO_cleaved), na.rm = TRUE),
    `KO+A` = mean(c(`KO+A_full`, `KO+A_cleaved`), na.rm = TRUE)
  ) %>%
  ungroup() %>%
  select(Gene, ProteinBase, WT, `WT+A`, KO, `KO+A`)
  
# Replace original rows with calculated means
# Perform the join
updated_data <- Protein.2 %>%
  left_join(means_data, by = c("Gene"))

# Replace original values with mean values where available
result <- updated_data %>%
  rowwise() %>%
  mutate(
    WT = coalesce(WT.y, WT.x),
    `WT+A` = coalesce(`WT+A.y`, `WT+A.x`),
    KO = coalesce(KO.y, KO.x),
    `KO+A` = coalesce(`KO+A.y`, `KO+A.x`)
  ) %>%
  ungroup() %>%
  select(Gene, Protein, WT, `WT+A`, KO, `KO+A`)

Protein.2 <- result %>%
  mutate(Protein = sub(" - .*", "", Protein)) %>%
  distinct(Gene, Protein, .keep_all = TRUE) 

# round to 3 sig figs
Protein.2 <- Protein.2 %>%
  mutate_if(is.numeric, round, 3)


Protein.2

This is the final processed data that calculates the mean for proteins that had multiple measurements. For clarity those proteins are AQP1, AQP2, ROMK, ENaC alpha, and ENaC gamma.

Cleaning the RNA data

Step 1: Rotate the dataframe.

#rotate df and create gene column 
RNA_cell <- Pseudobulk.cell %>%
  rotate_df() %>%
  mutate(Gene = rownames(.))

RNA_cell_group <- Pseudobulk.cell.group %>%
  rotate_df() %>%
  mutate(Gene = rownames(.))

RNA_group <- Pseudobulk.group %>%
  rotate_df() %>%
  mutate(Gene = rownames(.))

# round to 3 sig figs
RNA_group <- RNA_group %>%
  mutate_if(is.numeric, round, 3)

head(RNA_group)

In this step I rotated the data frame so the columns became rows and vice versa. I did this to make correlating the RNA and Protein data easier.

Step 2: Rename the colums

# Renaming the columns
RNA_group <- RNA_group %>% 
  dplyr::rename(
    WT = `Control Saline`,
    `WT+A` = `Control AngII`,
    KO = `PT ACE2 KO Saline`,
    `KO+A` = `PT ACE2 KO AngII`)

# round to 3 sig figs
RNA_group <- RNA_group %>%
  mutate_if(is.numeric, round, 3)

head(RNA_group)

In this step I renamed the columns so they matched the columns from the RNA data.

Step 3: Normalizing WT to 1

#gene_name <- unique(unlist(Protein.gene.data$Gene))
# Filter dataframe to keep only the selected genes
RNA_group <- RNA_group %>%
  filter(Gene %in% gene_name)

# Normalize the "WT" column to 1 and adjust other columns accordingly
RNA_group1 <- RNA_group %>%
  group_by(Gene) %>%
  mutate(
    `WT+A` = `WT+A` / WT[1],  
    KO = KO / WT[1],  
    `KO+A` = `KO+A` / WT[1],
    WT = WT / WT[1],)

# round to 3 sig figs
RNA_group1 <- RNA_group1 %>%
  mutate_if(is.numeric, round, 3)

RNA_group1

In this step I normalized the data frame so WT is normalized to 1, and other columns are proportionally adjusted based on this normalization. To do this I grouped by the Gene column to ensure calculations are performed separately for each gene. The I adjusted each by dividing its values by the first WT value within the group.

Visualize the data

Compare Protien/RNA data to orignal Western blot data

Original Protein Data

Figure 1 Bar graphs from Western blot data

Transformed Protein Data

# convert data frames into tidyform
Protein.2 <- Protein.2 %>%
  pivot_longer(cols = WT:`KO+A`, names_to = "Group", values_to = "Value")

# Reorder the X axis to be (WT, WT+A, KO, KO+A)
Protein.2$Group <- factor(Protein.2$Group, levels = c("WT", "WT+A", "KO", "KO+A"))

ggplot(Protein.2, aes(x = Gene, y = Value, fill = Group)) +
  geom_bar(stat = "identity", position = "dodge") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  labs(title = "Gene Expression Across Conditions from Protein data", x = "Gene", y = "Value") +
  scale_fill_brewer(palette = "Set1")

Transformed RNA Data

# convert data frames into tidyform
RNA_group1 <- RNA_group1 %>%
  pivot_longer(cols = WT:`KO+A`, names_to = "Group", values_to = "Value")

# Reorder the X axis to be (WT, WT+A, KO, KO+A)
RNA_group1$Group <- factor(RNA_group1$Group, levels = c("WT", "WT+A", "KO", "KO+A"))

ggplot(RNA_group1, aes(x = Gene, y = Value, fill = Group)) +
  geom_bar(stat = "identity", position = "dodge") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
  labs(title = "Gene Expression Across Conditions from RNA data", x = "Gene", y = "Value") +
  scale_fill_brewer(palette = "Set1")

Based on the bar graphs above, the transformed protein data displays similar patterns to the original protein data, for example Hmox1, Cldn10, and Ace2 have very similar expression patterns to the original data. However there is some variance in the RNA data.

Compare RNA/Protein data to orignal Psuedobulk data

Original RNA Data

# gene list
gene_name2 <- c("Ace", "Hmox1", "Slc5a2")

plots_f2 <- list()  # Initialize a list to store the plots

for (i in gene_name2) {
  if (i %in% colnames(Pseudobulk.group)) {
    selected_column <- Pseudobulk.group[, i, drop = FALSE]
    result_df2 <- data.frame(RowName = rownames(selected_column), Value = selected_column[, 1])
    result_df2$RowName <- factor(result_df2$RowName, levels = c("Control Saline", "Control AngII", "PT ACE2 KO Saline", "PT ACE2 KO AngII"))

    f2 <- ggplot(result_df2, aes(x = RowName, y = Value)) + 
      geom_boxplot() + 
      theme(axis.text.x = element_text(angle = 16, hjust = 1)) + 
      labs(title = paste(i, "- Original RNA Data"), x = "Experimental Groups", y = "Expression")
    
    plots_f2[[i]] <- f2  # Store the plot in the list
  }
}

Transformed RNA Data

# make col names Gene and row names Group 
RNA_group2 <- RNA_group1  %>% 
  select("Gene", "Group", "Value") %>%
  pivot_wider(names_from = Gene, values_from = Value) %>% 
  column_to_rownames(var = "Group")

plots_f3 <- list()  # Initialize a list to store the plots

for (i in gene_name2) {
  if (i %in% colnames(RNA_group2)) {
    selected_column <- RNA_group2[, i, drop = FALSE]
    result_df3 <- data.frame(RowName = rownames(selected_column), Value = selected_column[, 1])
    result_df3$RowName <- factor(result_df3$RowName, levels = c("WT", "WT+A", "KO", "KO+A"))

    f3 <- ggplot(result_df3, aes(x = RowName, y = Value)) + 
      geom_boxplot() + 
      theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
      labs(title = paste(i, "- Transformed RNA Data"), x = "Experimental Groups", y = "Expression")
    
    plots_f3[[i]] <- f3  # Store the plot in the list
  }
}

Transformed Protein Data

# make col names Gene and row names Group 
Protein.3 <- Protein.2  %>% 
  select("Gene", "Group", "Value") %>%
  pivot_wider(names_from = Gene, values_from = Value) %>% 
  column_to_rownames(var = "Group")

plots_f4 <- list()  # Initialize a list to store the plots

for (i in gene_name2) {
  if (i %in% colnames(Protein.3)) {
    selected_column <- Protein.3[, i, drop = FALSE]
    result_df4 <- data.frame(RowName = rownames(selected_column), Value = selected_column[, 1])
    result_df4$RowName <- factor(result_df4$RowName, levels = c("WT", "WT+A", "KO", "KO+A"))

    f4 <- ggplot(result_df4, aes(x = RowName, y = Value)) + 
      geom_boxplot() + 
      theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
      labs(title = paste(i, "- Transformed Protein Data"), x = "Experimental Groups", y = "Expression")
    
    plots_f4[[i]] <- f4  # Store the plot in the list
  }
}

Box plots comparing the dataframes

### Step 3: Combine and Display the Plots Side by Side

# Arrange plots for each gene side by side using cowplot's plot_grid
plot_list <- list()
for (i in gene_name2) {
  plot_list[[i]] <- plot_grid(plots_f2[[i]], plots_f3[[i]], plots_f4[[i]], ncol = 3)
}

# Display the combined plots for each gene
for (i in gene_name2) {
  print(plot_list[[i]])
}

Based on the box plots above, the transformed RNA data displays similar patterns to the original NRA data. Ace, Hmox1, and, Slc5a2 have the same expression patterns to the original data. However there is some variance in the protein data.

Scatter Plot

# get rid of protein column in protein data
Protein.2 <- Protein.2 %>%
  select(Gene, Group, Value)

# create Type column
RNA_group1$Type <- "RNA"

Protein.2$Type <- "Protein"

# Merge RNA and Protein df
combined_dataframe <- rbind(Protein.2, RNA_group1)

# Reorder the X axis to be (WT, WT+A, KO, KO+A)
combined_dataframe$Group <- factor(combined_dataframe$Group, levels = c("WT", "WT+A", "KO", "KO+A"))

# Create a scatter plot the matches the values of the Gene and Group column but splits them so that the x axis is from protein and the Y axis is from RNA
ggplot(combined_dataframe, aes(x = Group, y = Value, color = Type)) + 
  geom_point() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
  facet_wrap(~Gene, scales = "free_y") +   
  labs(title = "Scatter Plot of Value by Group and Type", x = "Group", y = "Value")

#same graph with fixed axis
ggplot(combined_dataframe, aes(x = Group, y = Value, color = Type)) + 
  geom_point() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
  facet_wrap(~Gene, scales = "free_y") + 
  coord_cartesian(ylim = c(0.3, 2.2)) +   
  labs(title = "Scatter Plot of Value by Group and Type: Fixed Axis", x = "Group", y = "Value")

Basic Correlation

# Create a new dataframe that creates new columns to have the values for RNA and Protein in the same row
combined_dataframe2 <- combined_dataframe %>%
  pivot_wider(names_from = Type, values_from = Value)

# Create a new column if RNA is greater than 1
combined_dataframe2$RNA_Greater_1 <- ifelse(combined_dataframe2$RNA > 1, "Yes", "No")
combined_dataframe2$Protein_Greater_1 <- ifelse(combined_dataframe2$Protein > 1, "Yes", "No")

#create a  column if RNA_Greater_1 and Protein_Greater_1 are the same value
combined_dataframe2$Same <- ifelse(combined_dataframe2$RNA_Greater_1 == combined_dataframe2$Protein_Greater_1, "Yes", "No")

# Count how many values are yes or no in the column Same
combined_dataframe2 %>%
  group_by(Same) %>%
  summarise(Count = n())

#combined_dataframe2 %>%
  #group_by(Gene, Same) %>%
  #summarise(Count = n())

This table helps assess the correlation between RNA and protein expression levels. It takes into account whether RNA and protein levels are greater than 1, then counts how often these conditions match.

Heat maps of the change within groups

Transformed data

# create absolute value of difference
abs_diff_df3 <- combined_dataframe %>%
  pivot_wider(names_from = Type, values_from = Value) %>%
  mutate(abs_diff = abs(RNA - Protein))

#give column names + sum all deltas
abs_diff_df3 <- abs_diff_df3 %>%
  group_by(Gene) %>%
  summarise(
    abs_diff_wt = abs_diff[Group == "WT"],
    abs_diff_ko = abs_diff[Group == "KO"],
    abs_diff_wt_a = abs_diff[Group == "WT+A"],
    abs_diff_ko_a = abs_diff[Group == "KO+A"]
  ) %>%
  mutate(sum_delta = abs_diff_wt + abs_diff_ko + abs_diff_wt_a+ abs_diff_ko_a)

#same graph with fixed axis
ggplot(combined_dataframe, aes(x = Group, y = Value, color = Type)) + 
  geom_point() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
  facet_wrap(~Gene, scales = "free_y") + 
  coord_cartesian(ylim = c(0.3, 2.2)) +   
  labs(title = "Scatter Plot of Value by Group and Type: Fixed Axis", x = "Group", y = "Value")

WT vs KO

# subset WT and KO delete once u figure out cor test
combined_WT_KO <-subset(combined_dataframe, Group %in% c("WT", "KO"))

# Calculate the absolute difference between RNA and Protein values
abs_diff_WT_KO <- combined_WT_KO %>%
  pivot_wider(names_from = Type, values_from = Value) %>%
  mutate(abs_diff = abs(RNA - Protein))

#give column names + sum all deltas 
abs_diff_WT_KO <- abs_diff_WT_KO %>%
  group_by(Gene) %>%
  summarise(
    abs_diff_wt = abs_diff[Group == "WT"],
    abs_diff_ko = abs_diff[Group == "KO"],
  ) %>%
  mutate(sum_delta = abs_diff_wt + abs_diff_ko)

head(abs_diff_WT_KO)

#same graph with fixed axis
ggplot(combined_WT_KO, aes(x = Group, y = Value, color = Type)) + 
  geom_point() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
  facet_wrap(~Gene, scales = "free_y") + 
  coord_cartesian(ylim = c(0, 2)) +   
  labs(title = "Scatter Plot of Value by Group and Type: WT vs KO", x = "Group", y = "Value")

WT vs WT+A

# subset WT and WT+A
combined_WT_WTA <-subset(combined_dataframe, Group %in% c("WT", "WT+A"))

# Calculate the absolute difference between RNA and Protein values
abs_diff_WT_WTA <- combined_WT_WTA %>%
  pivot_wider(names_from = Type, values_from = Value) %>%
  mutate(abs_diff = abs(RNA - Protein))


#give column names + sum all deltas
abs_diff_WT_WTA <- abs_diff_WT_WTA %>%
  group_by(Gene) %>%
  summarise(
    abs_diff_wt = abs_diff[Group == "WT"],
    abs_diff_wt_a = abs_diff[Group == "WT+A"],
  ) %>% 
  mutate(
    sum_delta = abs_diff_wt + abs_diff_wt_a)

head(abs_diff_WT_WTA)

ggplot(combined_WT_WTA, aes(x = Group, y = Value, color = Type)) + 
  geom_point() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
  facet_wrap(~Gene, scales = "free_y") + 
  coord_cartesian(ylim = c(0.4, 1.8)) +   
  labs(title = "Scatter Plot of Value by Group and Type: WT vs WT+A", x = "Group", y = "Value")

KO vs KO+A

# subset KO and KO+A
combined_KO_KOA <-subset(combined_dataframe, Group %in% c("KO", "KO+A"))

# create separate columns for "WT" and "KO" values
pivoted_KO_KOA <- combined_KO_KOA %>%
  pivot_wider(names_from = Group, values_from = Value)

# normalize KO to 1 
pivoted_KO_KOA <- pivoted_KO_KOA %>%
  group_by(Gene, Type) %>%
  mutate(
    `KO+A` = `KO+A` / KO[1],
    KO = KO / KO[1],) %>%
  ungroup()

# recombine columns  
combined_KO_KOA <- pivoted_KO_KOA %>%
  pivot_longer(cols = c(KO, `KO+A`), names_to = "Group", values_to = "Value")

# Calculate the absolute difference between RNA and Protein values
abs_diff_KO_KOA  <- combined_KO_KOA %>%
  pivot_wider(names_from = Type, values_from = Value) %>%
  mutate(abs_diff = abs(RNA - Protein))

#give column names + sum all deltas
abs_diff_KO_KOA <- abs_diff_KO_KOA %>%
  group_by(Gene) %>%
  summarise(
    abs_diff_ko = abs_diff[Group == "KO"],
    abs_diff_ko_a = abs_diff[Group == "KO+A"]
  ) %>%
  mutate(sum_delta = abs_diff_ko + abs_diff_ko_a)

# round to 3 sig figs
abs_diff_KO_KOA <- abs_diff_KO_KOA %>%
  mutate_if(is.numeric, round, 3)

head(abs_diff_KO_KOA)

ggplot(combined_KO_KOA, aes(x = Group, y = Value, color = Type)) + 
  geom_point() + 
  theme(axis.text.x = element_text(angle = 90, hjust = 1)) + 
  facet_wrap(~Gene, scales = "free_y") + 
  coord_cartesian(ylim = c(0.4, 1.4)) +   
  labs(title = "Scatter Plot of Value by Group and Type: KO vs KO+A", x = "Group", y = "Value")

Heat maps

abs_diff_KO_KOA$Metric <- "KO vs KO+A"
abs_diff_WT_WTA$Metric <- "WT vs WT+A"
abs_diff_df3$Metric <- "Transformed"
abs_diff_WT_KO$Metric <- "WT vs WT+KO"

abs_diff_hm_combined <- bind_rows(
  mutate(abs_diff_KO_KOA, Metric = "KO vs KO+A"),
  mutate(abs_diff_WT_WTA, Metric = "WT vs WT+A"),
  mutate(abs_diff_WT_KO, Metric = "WT vs KO"),
  mutate(abs_diff_df3, Metric = "Transformed")
)

# order metrics
abs_diff_hm_combined$Metric <- factor(abs_diff_hm_combined$Metric, levels = c("Transformed", "WT vs KO", "WT vs WT+A", "KO vs KO+A"))

# Create the heat map with facets
ggplot(abs_diff_hm_combined, aes(x = Metric, y = Gene, fill = sum_delta)) +
  geom_tile() +
  scale_fill_viridis(option = "magma") +
  facet_wrap(~ Metric, scales = "free_x", ncol = 4) +  # Line up metrics horizontally
  theme_minimal() +
  theme(
    axis.text.x = element_blank(),    # Remove x-axis text
    axis.ticks.x = element_blank(),   # Remove x-axis ticks
    strip.text = element_text(size = 12, angle = 0),  # Adjust facet label size and angle if needed
    panel.spacing = unit(1, "lines") # Space between facets
  ) +
  labs(
    title = "Sum of |Delta| Heat Maps",
    x = "",  # Remove x-axis label as it is redundant
    y = "Gene/Protein",
    fill = "|Delta|"
  )

# separate dynamic range for transformed data
# Generate the shared dynamic range heat maps
shared_dynamic_range_df <- bind_rows(
  mutate(abs_diff_WT_KO, Metric = "WT vs KO"),
  mutate(abs_diff_WT_WTA, Metric = "WT vs WT+A"),
  mutate(abs_diff_KO_KOA, Metric = "KO vs KO+A")
)

# order metrics
shared_dynamic_range_df$Metric <- factor(shared_dynamic_range_df$Metric, levels = c("Transformed", "WT vs KO", "WT vs WT+A", "KO vs KO+A"))

# Generate the transformed heat map separately
transformed_dynamic_range_df <- mutate(abs_diff_df3, Metric = "Transformed")

# Common title for the plots
common_title <- "Sum of |Delta| Heat Maps"

# Plot the shared dynamic range heat maps
shared_hm <- ggplot(shared_dynamic_range_df, aes(x = Metric, y = Gene, fill = sum_delta)) +
  geom_tile() +
  scale_fill_viridis(option = "magma") +
  theme_minimal() +
  theme(
    axis.text.x = element_blank(),    # Remove x-axis text
    axis.ticks.x = element_blank(),   # Remove x-axis ticks
    strip.text = element_text(size = 12, angle = 0),  # Adjust facet label size and angle if needed
    panel.spacing = unit(1, "lines"), # Space between facets
    panel.grid = element_blank()      # Remove all grid marks
  ) +
  labs(
    title = "",
    x = "",  # Remove x-axis label as it is redundant
    y = "Gene/Protein",
    fill = "|Delta|"
  ) +
  facet_wrap(~ Metric, scales = "free_x", ncol = 3)

# Plot the transformed heat map
transformed_hm <- ggplot(transformed_dynamic_range_df, aes(x = Metric, y = Gene, fill = sum_delta)) +
  geom_tile() +
  scale_fill_viridis(option = "magma") +
  coord_fixed(ratio = .3) +
  theme_minimal() +
  theme(
    axis.text.x = element_blank(),    # Remove x-axis text
    axis.ticks.x = element_blank(),   # Remove x-axis ticks
    strip.text = element_text(size = 12, angle = 0),  # Adjust facet label size and angle if needed
    panel.spacing = unit(1, "lines"), # Space between facets
    panel.grid = element_blank()      # Remove all grid marks
  ) +
  labs(
    title = "",
    x = "Transformed",   
    y = "Gene/Protein",
    fill = "|Delta|"
  )

# Arrange the heat maps side by side
grid.arrange(transformed_hm, shared_hm, ncol = 2, top = common_title)

Bar graph of Sum of |Delta|

# Calculate the mean and standard error of the points
hm_summary <- abs_diff_hm_combined %>%
  group_by(Metric) %>%
  summarize(
    mean_sum_delta = mean(sum_delta),
    SE = sd(sum_delta) / sqrt(n()))

# Create the scatter plot with mean outlined as a bar graph + SE bars 
ggplot(abs_diff_hm_combined, aes(x = Metric, y = sum_delta)) +
  geom_jitter(width = 0.1, height = 0, alpha = 0.6) +  #  better visibility
geom_errorbar(data = hm_summary, aes(x = factor(Metric), y = mean_sum_delta, ymin = mean_sum_delta - SE, ymax = mean_sum_delta + SE), width = 0.2, color = "black") +  # Standard error bars
    geom_col(data = hm_summary, aes(y = mean_sum_delta), fill = NA, color = "black", size = 1) +  # Outline of bar for the mean
  theme_minimal() +
  labs(title = "Scatter Plot with Mean Outline and Standard Error Bars",
       x = "Metric",
       y = "Sum_Delta")

# table of sum of sum_delta for each metric
hm_sum <- abs_diff_hm_combined %>%
  group_by(Metric) %>%
  summarize(Sum = sum(sum_delta))

# round to 3 sig figs
hm_sum <- hm_sum %>%
  mutate_if(is.numeric, round, 3)

hm_sum

The table above represents the sum of each of the delta values for each condition.

T test

WT/KO vs WT/WT+A

t_test_df_1 <- abs_diff_WT_KO %>%
  merge(abs_diff_WT_WTA, by = "Gene")

t_test <- t.test(t_test_df_1$sum_delta.x, t_test_df_1$sum_delta.y)
t_test

## 
##  Welch Two Sample t-test
## 
## data:  t_test_df_1$sum_delta.x and t_test_df_1$sum_delta.y
## t = -0.13855, df = 33.126, p-value = 0.8906
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.1857076  0.1620234
## sample estimates:
## mean of x mean of y 
## 0.3066316 0.3184737

WT/WT+A vs KO/KO+A

t_test_df_2 <- abs_diff_WT_WTA %>%
  merge(abs_diff_KO_KOA, by = "Gene")

t_test2 <- t.test(t_test_df_2$sum_delta.x, t_test_df_2$sum_delta.y)
t_test2

## 
##  Welch Two Sample t-test
## 
## data:  t_test_df_2$sum_delta.x and t_test_df_2$sum_delta.y
## t = 1.5022, df = 28.882, p-value = 0.1439
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.0431957  0.2820378
## sample estimates:
## mean of x mean of y 
## 0.3184737 0.1990526

WT/KO vs KO/KO+A

t_test_df_3 <- abs_diff_WT_KO %>%
  merge(abs_diff_KO_KOA, by = "Gene")

t_test3 <- t.test(t_test_df_3$sum_delta.x, t_test_df_3$sum_delta.y)
t_test3

## 
##  Welch Two Sample t-test
## 
## data:  t_test_df_3$sum_delta.x and t_test_df_3$sum_delta.y
## t = 1.6664, df = 34.094, p-value = 0.1048
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.02360552  0.23876341
## sample estimates:
## mean of x mean of y 
## 0.3066316 0.1990526

Heat map of R-squared Values

Transformed data

# Group by Gene and calculate R-squared for each gene
r_squared_values <- combined_dataframe %>%
  group_by(Gene) %>%
  summarize(
    R_squared = summary(lm(Value ~ Group))$r.squared
  ) %>%
  ungroup()

# Create a data frame for plotting
r_squared_df <- r_squared_values %>%
  arrange(desc(R_squared)) %>%
  mutate(Gene = factor(Gene, levels = Gene))

# round to 3 sig figs
r_squared_values <- r_squared_values %>%
  mutate_if(is.numeric, round, 3)

head(r_squared_values)

WT vs KO

# Group by Gene and calculate R-squared for each gene
r_squared_values_WT_KO <- combined_WT_KO %>%
  group_by(Gene) %>%
  summarize(
    R_squared = summary(lm(Value ~ Group))$r.squared
  ) %>%
  ungroup()

# Create a data frame for plotting
r_squared_df_WT_KO <- r_squared_values_WT_KO %>%
  arrange(desc(R_squared)) %>%
  mutate(Gene = factor(Gene, levels = Gene))

# round to 3 sig figs
r_squared_values_WT_KO <- r_squared_values_WT_KO %>%
  mutate_if(is.numeric, round, 3)

head(r_squared_values_WT_KO)

WT vs WT+A

# WT vs WT+A
# Group by Gene and calculate R-squared for each gene
r_squared_values_WT_WTA <- combined_WT_WTA %>%
  group_by(Gene) %>%
  summarize(
    R_squared = summary(lm(Value ~ Group))$r.squared
  ) %>%
  ungroup()

# Create a data frame for plotting
r_squared_df_WT_WTA <- r_squared_values_WT_WTA %>%
  arrange(desc(R_squared)) %>%
  mutate(Gene = factor(Gene, levels = Gene))

# round to 3 sig figs
r_squared_values_WT_WTA <- r_squared_values_WT_WTA %>%
  mutate_if(is.numeric, round, 3)

head(r_squared_values_WT_WTA)

KO vs KO+A

# Group by Gene and calculate R-squared for each gene
r_squared_values_KO_KOA <- combined_KO_KOA %>%
  group_by(Gene) %>%
  summarize(
    R_squared = summary(lm(Value ~ Group))$r.squared
  ) %>%
  ungroup()

# Create a data frame for plotting
r_squared_df_KO_KOA <- r_squared_values_KO_KOA %>%
  arrange(desc(R_squared)) %>%
  mutate(Gene = factor(Gene, levels = Gene))

# round to 3 sig figs
r_squared_values_KO_KOA <- r_squared_values_KO_KOA %>%
  mutate_if(is.numeric, round, 3)

head(r_squared_values_KO_KOA)

Heat maps

r_squared_df_KO_KOA$Metric <- "KO vs KO+A"
r_squared_df_WT_WTA$Metric <- "WT vs WT+A"
r_squared_df$Metric <- "Transformed"
r_squared_df_WT_KO$Metric <- "WT vs WT+KO"

hm_combined_df <- bind_rows(
  mutate(r_squared_df_KO_KOA, Metric = "KO vs KO+A"),
  mutate(r_squared_df_WT_WTA, Metric = "WT vs WT+A"),
  mutate(r_squared_df_WT_KO, Metric = "WT vs KO"),
  mutate(r_squared_df, Metric = "Transformed")
)

# Create the heat map with facets
hm_combined_df$Metric <- factor(hm_combined_df$Metric, levels = c("Transformed", "WT vs KO", "WT vs WT+A", "KO vs KO+A"))

ggplot(hm_combined_df, aes(x = Metric, y = Gene, fill = R_squared)) +
  geom_tile() +
  scale_fill_viridis(option = "magma") +
  facet_wrap(~ Metric, scales = "free_x", ncol = 4) +  # Line up metrics horizontally
  theme_minimal() +
  theme(
    axis.text.x = element_blank(),    # Remove x-axis text
    axis.ticks.x = element_blank(),   # Remove x-axis ticks
    strip.text = element_text(size = 12, angle = 0),  # Adjust facet label size and angle if needed
    panel.spacing = unit(1, "lines") # Space between facets
  ) +
  labs(
    title = "R-squared Values Heat Maps",
    x = "",  # Remove x-axis label as it is redundant
    y = "Gene/Protein",
    fill = "R-squared"
  )

# separate dynamic range for transformed data
# Generate the shared dynamic range heat maps
shared_dynamic_range_df2 <- bind_rows(
  mutate(r_squared_df_KO_KOA, Metric = "KO vs KO+A"),
  mutate(r_squared_df_WT_WTA, Metric = "WT vs WT+A"),
  mutate(r_squared_df_WT_KO, Metric = "WT vs KO")
)

shared_dynamic_range_df2$Metric <- factor(shared_dynamic_range_df2$Metric, levels = c("WT vs KO", "WT vs WT+A", "KO vs KO+A"))

# Generate the transformed heat map separately
transformed_dynamic_range_df2 <- mutate(r_squared_df, Metric = "Transformed")

# Common title for the plots
common_title2 <- "R-Squared Heat Maps"

# Plot the shared dynamic range heat maps
shared_hm2 <- ggplot(shared_dynamic_range_df2, aes(x = Metric, y = Gene, fill = R_squared)) +
  geom_tile() +
  scale_fill_viridis(option = "magma") +
  theme_minimal() +
  theme(
    axis.text.x = element_blank(),    # Remove x-axis text
    axis.ticks.x = element_blank(),   # Remove x-axis ticks
    strip.text = element_text(size = 12, angle = 0),  # Adjust facet label size and angle if needed
    panel.spacing = unit(1, "lines") # Space between facets
  ) +
  labs(
    title = "",
    x = "",  # Remove x-axis label as it is redundant
    y = "Gene/Protein",
    fill = "R-Sqaured"
  ) +
  facet_wrap(~ Metric, scales = "free_x", ncol = 3)

# Plot the transformed heat map
transformed_hm2 <- ggplot(transformed_dynamic_range_df2, aes(x = Metric, y = Gene, fill = R_squared)) +
  geom_tile() +
  scale_fill_viridis(option = "magma") +
  coord_fixed(ratio = .3) +
  theme_minimal() +
  theme(
    axis.text.x = element_blank(),    # Remove x-axis text
    axis.ticks.x = element_blank(),   # Remove x-axis ticks
    strip.text = element_text(size = 12, angle = 0),  # Adjust facet label size and angle if needed
    panel.spacing = unit(1, "lines") # Space between facets
  ) +
  labs(
    title = "",
    x = "Transformed",   
    y = "Gene/Protein",
    fill = "R Squared"
  )

# Arrange the heat maps side by side
grid.arrange(transformed_hm2, shared_hm2, ncol = 2, top = common_title2)

Ratio between the mean absolute values of the differnce in AngII groups and the differnce in the KO group

# Calculate the absolute difference between RNA and Protein values
abs_diff_df <- combined_dataframe %>%
  pivot_wider(names_from = Type, values_from = Value) %>%
  mutate(abs_diff = abs(RNA - Protein))

# Calculate the mean absolute difference for "wt+a" and "ko+a" groups separately for each gene
#make new column that represents the average of the wt+a and ko+a conditions (mean_a)
#make new column that represents the ratio between the mean_a and ko conditions (ratio)
abs_diff_df2 <- abs_diff_df %>%
  group_by(Gene) %>%
  summarise(
    abs_diff_wt = abs_diff[Group == "WT"],
    abs_diff_ko = abs_diff[Group == "KO"],
    abs_diff_wt_a = abs_diff[Group == "WT+A"],
    abs_diff_ko_a = abs_diff[Group == "KO+A"]
  ) %>%
  mutate(mean_a = (abs_diff_ko_a + abs_diff_wt_a) /2) %>%
  mutate(ratio = mean_a/abs_diff_ko) 

# arrange so the df is in order of descending log_ratio
abs_diff_df2 <- abs_diff_df2 %>%
  mutate(log_ratio = log10(ratio)) %>%  
  arrange(desc(log_ratio)) %>%   
  mutate(Gene = factor(Gene, levels = unique(Gene)))   

# Plot with genes in descending order by log_ratio
ggplot(abs_diff_df2, aes(x = Gene, y = ratio)) +
  geom_bar(stat = "identity", fill = "navy") +  
  scale_y_log10() +
  labs(title = "Ratio of AngII Conditions and the KO Condition",
       x = "Gene/Protein",
       y = "Ratio: ((WT+A)+ (KO+A)/ KO)") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 90, hjust = 1))

Session Info

sessionInfo()

## R version 4.4.0 (2024-04-24)
## Platform: x86_64-apple-darwin20
## Running under: macOS Sonoma 14.6.1
## 
## Matrix products: default
## BLAS:   /Library/Frameworks/R.framework/Versions/4.4-x86_64/Resources/lib/libRblas.0.dylib 
## LAPACK: /Library/Frameworks/R.framework/Versions/4.4-x86_64/Resources/lib/libRlapack.dylib;  LAPACK version 3.12.0
## 
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
## 
## time zone: America/Los_Angeles
## tzcode source: internal
## 
## attached base packages:
## [1] grid      stats4    stats     graphics  grDevices utils     datasets 
## [8] methods   base     
## 
## other attached packages:
##  [1] cowplot_1.1.3               viridis_0.6.5              
##  [3] viridisLite_0.4.2           pheatmap_1.0.12            
##  [5] ggvenn_0.1.10               gplots_3.1.3.1             
##  [7] gridExtra_2.3               sjmisc_2.8.10              
##  [9] devtools_2.4.5              usethis_2.2.3              
## [11] reshape2_1.4.4              here_1.0.1                 
## [13] data.table_1.15.4           ggpmisc_0.6.0              
## [15] ggpp_0.5.8-1                gghighlight_0.4.1          
## [17] readxl_1.4.3                openxlsx_4.2.5.2           
## [19] car_3.1-2                   carData_3.0-5              
## [21] RColorBrewer_1.1-3          lubridate_1.9.3            
## [23] forcats_1.0.0               purrr_1.0.2                
## [25] readr_2.1.5                 tidyr_1.3.1                
## [27] tibble_3.2.1                tidyverse_2.0.0            
## [29] DESeq2_1.44.0               SummarizedExperiment_1.34.0
## [31] Biobase_2.64.0              MatrixGenerics_1.16.0      
## [33] matrixStats_1.3.0           GenomicRanges_1.56.1       
## [35] GenomeInfoDb_1.40.1         IRanges_2.38.1             
## [37] S4Vectors_0.42.1            BiocGenerics_0.50.0        
## [39] EnhancedVolcano_1.22.0      ggrepel_0.9.5              
## [41] ggplot2_3.5.1               patchwork_1.2.0            
## [43] glmGamPoi_1.16.0            sctransform_0.4.1          
## [45] Seurat_5.1.0                SeuratObject_5.0.2         
## [47] sp_2.1-4                    stringr_1.5.1              
## [49] dplyr_1.1.4                 BiocManager_1.30.23        
## [51] knitr_1.48                 
## 
## loaded via a namespace (and not attached):
##   [1] RcppAnnoy_0.0.22        splines_4.4.0           later_1.3.2            
##   [4] bitops_1.0-7            cellranger_1.1.0        polyclip_1.10-6        
##   [7] fastDummies_1.7.3       lifecycle_1.0.4         rprojroot_2.0.4        
##  [10] globals_0.16.3          lattice_0.22-6          MASS_7.3-61            
##  [13] insight_0.20.2          magrittr_2.0.3          plotly_4.10.4          
##  [16] sass_0.4.9              rmarkdown_2.27          remotes_2.5.0          
##  [19] jquerylib_0.1.4         yaml_2.3.9              httpuv_1.6.15          
##  [22] spam_2.10-0             zip_2.3.1               sessioninfo_1.2.2      
##  [25] pkgbuild_1.4.4          spatstat.sparse_3.1-0   reticulate_1.38.0      
##  [28] pbapply_1.7-2           pkgload_1.4.0           abind_1.4-5            
##  [31] zlibbioc_1.50.0         Rtsne_0.17              GenomeInfoDbData_1.2.12
##  [34] irlba_2.3.5.1           listenv_0.9.1           spatstat.utils_3.0-5   
##  [37] MatrixModels_0.5-3      goftest_1.2-3           RSpectra_0.16-1        
##  [40] spatstat.random_3.3-1   fitdistrplus_1.2-1      parallelly_1.37.1      
##  [43] leiden_0.4.3.1          codetools_0.2-20        DelayedArray_0.30.1    
##  [46] tidyselect_1.2.1        farver_2.1.2            UCSC.utils_1.0.0       
##  [49] spatstat.explore_3.3-1  jsonlite_1.8.8          ellipsis_0.3.2         
##  [52] progressr_0.14.0        ggridges_0.5.6          survival_3.7-0         
##  [55] tools_4.4.0             ica_1.0-3               Rcpp_1.0.13            
##  [58] glue_1.7.0              SparseArray_1.4.8       xfun_0.46              
##  [61] withr_3.0.0             fastmap_1.2.0           fansi_1.0.6            
##  [64] SparseM_1.84-2          caTools_1.18.2          digest_0.6.36          
##  [67] timechange_0.3.0        R6_2.5.1                mime_0.12              
##  [70] colorspace_2.1-0        scattermore_1.2         gtools_3.9.5           
##  [73] tensor_1.5              spatstat.data_3.1-2     utf8_1.2.4             
##  [76] generics_0.1.3          httr_1.4.7              htmlwidgets_1.6.4      
##  [79] S4Arrays_1.4.1          uwot_0.2.2              pkgconfig_2.0.3        
##  [82] gtable_0.3.5            lmtest_0.9-40           XVector_0.44.0         
##  [85] htmltools_0.5.8.1       profvis_0.3.8           dotCall64_1.1-1        
##  [88] scales_1.3.0            png_0.1-8               spatstat.univar_3.0-0  
##  [91] rstudioapi_0.16.0       tzdb_0.4.0              nlme_3.1-165           
##  [94] cachem_1.1.0            zoo_1.8-12              sjlabelled_1.2.0       
##  [97] KernSmooth_2.23-24      parallel_4.4.0          miniUI_0.1.1.1         
## [100] pillar_1.9.0            vctrs_0.6.5             RANN_2.6.1             
## [103] urlchecker_1.0.1        promises_1.3.0          xtable_1.8-4           
## [106] cluster_2.1.6           evaluate_0.24.0         cli_3.6.3              
## [109] locfit_1.5-9.10         compiler_4.4.0          rlang_1.1.4            
## [112] crayon_1.5.3            future.apply_1.11.2     labeling_0.4.3         
## [115] fs_1.6.4                plyr_1.8.9              stringi_1.8.4          
## [118] deldir_2.0-4            BiocParallel_1.38.0     munsell_0.5.1          
## [121] lazyeval_0.2.2          spatstat.geom_3.3-2     quantreg_5.98          
## [124] Matrix_1.7-0            RcppHNSW_0.6.0          hms_1.1.3              
## [127] future_1.33.2           shiny_1.8.1.1           highr_0.11             
## [130] ROCR_1.0-11             memoise_2.0.1           igraph_2.0.3           
## [133] bslib_0.7.0             polynom_1.4-1

RNA-Protein Correlation

Sienna Blanche

26 August 2024

Load Packages

Project Narritive

Protein Processing

Narrative of workflow

Important notes

Interpretation

Load Data

Visualize Genes of Interest

Cleaning the Protein data

Step 1: Clean the raw data

Step 2: Calculate the weighted average for cortex and medullary proteins

Step 3: Calculate the average for proteins with multiple measurements

Cleaning the RNA data

Step 1: Rotate the dataframe.

Step 2: Rename the colums

Step 3: Normalizing WT to 1

Visualize the data

Compare Protien/RNA data to orignal Western blot data

Original Protein Data

Transformed Protein Data

Transformed RNA Data

Compare RNA/Protein data to orignal Psuedobulk data

Original RNA Data

Transformed RNA Data

Transformed Protein Data

Box plots comparing the dataframes

Scatter Plot

Basic Correlation

Heat maps of the change within groups

Transformed data

WT vs KO

WT vs WT+A

KO vs KO+A

Heat maps

Bar graph of Sum of |Delta|

T test

WT/KO vs WT/WT+A

WT/WT+A vs KO/KO+A

WT/KO vs KO/KO+A

Heat map of R-squared Values

Transformed data

WT vs KO

WT vs WT+A

KO vs KO+A

Heat maps

Ratio between the mean absolute values of the differnce in AngII groups and the differnce in the KO group

Session Info